Application Of Random Forest And Data Visualization In Prediction Of Cotton Aphid Grade

The monitoring and early warning of cotton aphid is the focus of the research on the prevention and control of cotton aphid.The relevant data of cotton aphid are collected and analyzed,and the control of cotton aphid is carried out in advance to reduce the harm caused by cotton aphid and cotton.Production.The research process of data analysis is carried out from two aspects: one is from the data visualization point of view of the data display analysis;the second is the use of high-performance machine algorithm.In this paper,we use the random forest algorithm to analyze the data of cotton aphid.Random forest is composed of multiple decision trees which constitute an integrated classification machine learning algorithm,used to predict the data classification.Decision trees and multivariate linear regression algorithms are also commonly used to predict data as Random Forests.However,the difference of the algorithm may lead to the fact that same data set on the prediction rate is inconsistent,so this paper on the three algorithms in the UCI dataset and the worms on the accuracy of the experimental data on the experiment.At present,the Linear Regression model of cotton aphid pest prediction is used.The shortcoming of the Linear Regression model is that the expression of the factor is only a kind of speculation,which affects the diversity and unmeasurable nature of the factor.The construction of the Random Forest model will not be affected by the expression of the influencing factors.Moreover,the Random Forest algorithm will not produce the fitting,the processing time of the large sample set is fast,and the accuracy of the classification prediction is high.In the previous experiment,the experiment shows that the Random Forest has high accuracy in the data forecasting.The later experiment is based on the Random Forest algorithmin the prediction of cotton aphid grade.Cotton is an important economic crop in our country,and it plays a huge role in the agricultural economic pattern.Aphis gossypii is the main factor causing cotton production and yield,so the advance prevention and control of cotton aphid is very important.In this paper,the data collected on the imbalance processing and the impact of factors after screening,and constructs a random forest model based on meteorological factor data and natural enemy data of Aphis gossypii,and uses the constructed model to predict the occurrence of cotton aphids.This experiment shows that the generalization error of the Random Forest model is small,and the accuracy rate of the cotton aphid rank is higher.Second,the use of data visualization technology for data analysis.Data visualization is an important means of data analysis at present,and ultidimensional data visualization is one of the focuses of data visualization.Through the multi-dimensional data display,found that the link between attributes.Currently,we collected the data for the multi-dimensional data,the collected meteorological data and cotton aphid data visual display,found that data hidden information,help to better data analysis and decision-making.The presentation and analysis of the data in this paper makes it possible to understand the time of occurrence of cotton aphid in order to provide reference for prevention and control.The visualization of the data in the experiment plays an important role in the construction of the model and the demonstration and analysis of the experimental results.